ADVANCED MATERIALS & PROCESSES | NOVEMBER/DECEMBER 2024 18 steps and mechanical methods due to errors in strain gage readings. Thin or thick section; the stress gradient can be large in thin parts and methods that can capture steep gradients should be employed. Accuracy vs. precision of residual stress measurement. Typically, with all measurement methods, there is a tradeoff between time/cost and accuracy/precision of the given measurement. Process optimization and quality control would need precision over accuracy, where model calibration, life, and design tasks would need accuracy over precision. Material considerations. Mechanical relaxation methods can be used on all materials regardless of crystallinity state; however, in materials with very high or very low yield strength, cutting methods can introduce artifacts into the data. Diffraction-based methods work only on crystalline materials. Raman methods are applicable typically on materials that show a Raman shift, such as ceramics. Most methods assume either a near-random grain orientation or a negligible crystallographic texture; therefore, materials exhibiting either a significant crystallo- graphic texture or are composed of a single crystal must be treated using suitable assumptions. Standard measurement techniques can be applied for isotropic, polycrystalline materials with small crystallite size (<100 μm) and low texture (<3-4 times random). In the case of materials with high anisotropy, or with high texture, or single crystal materials, or materials with grains larger than 100 μm, specialized methods that take into consideration anisotropy must be applied[1]. Stress tensor components. It is important to ask what components of the stress tensor, or types of stress, are required for a given application. Often only one or two stress components are measured. In these cases, assumptions are made when converting strains to stresses using Hooke’s law. It is important to ensure any assumptions are applicable for the given case. Sampling volume. Sampling volume is defined by the data application and subsequently the selected method and associated settings. It is not uncommon to get varying residual stress results when choosing different sampling volumes or methods where sampling volumes are different because if a stress gradient is present in the part, the “stress value” is location dependent. Residual stress predictive model- ing tools. The measurement resources can be focused and guided using residual stress predictive modeling tools as a precursor to a measurement plan definition. This enables agile response and iterative calibration and validation of available models resulting in a comprehensive understating of residual stress fields by location and stress component. Further, after a model is calibrated, it can be used for the given application space without or with very few additional measurements. SUMMARY A wealth of measurement techniques are available to materials scientists. It is critical to understand data TABLE 3 — ULTRASOUND AND MAGNETIC METHODS FOR RESIDUAL STRESS MEASUREMENT Ultrasound methods Magnetic methods Method Acoustic-elasticity Magnet-inductive Magneto-elastic Ferro-magnetic What is measured? Velocities of sound waves Permeability Amplitude of the magnetostrictively excited ultrasonic wave Amplitude of magnetic Barkhausen noise Smallest probing area ~30 mm2 ~1 mm2 ~10 mm2 ~0.5 mm2 Special lateral resolution 5 mm 1 mm 1 mm 1 mm Depth achievable nondestructively >10 cm ~1 cm ~1 cm ~1 cm Stress types Type I Type I Type I Type I Stress state measurable Uniaxial, biaxial Uniaxial, biaxial Uniaxial, biaxial Uniaxial, biaxial Measurable materials Crystalline metals, ceramics Crystalline ferro-magnetic materials Crystalline ferro-magnetic materials Crystalline ferro-magnetic materials Accuracy ±10 MPa ±10 MPa ±10 MPa ±10 MPa Precision ±10 MPa 10% 10% 10%
RkJQdWJsaXNoZXIy MTYyMzk3NQ==